.svg){kind=logo}
MARKOV SWITCHING GENERALIZED AUTOREGRESSIVE CONDITIONAL HETEROSKEDASTICITY (MSGARCH) MODEL BASED ON LARGE DEVIATION PRINCIPLE
Keywords:
heteroskedasticity, variance regime, hidden Markov model, large deviation principle, regime switching, MSGARCH, LDP-MSGARCHDOI:
https://doi.org/10.17654/0972086325015Abstract
Variance breaks in time series data result in heteroskedasticity, influencing the distribution and modeling of the series. Volatility models, like as GARCH and ARCH, have exhibited constraints in accurately modeling this type of time series data. The constraints of current models have prompted the creation of regime-switching volatility models, which have proven effective in modeling such time series. Markov switching volatility models provide improved efficiency by incorporating regime dependencies in the examination of regime transitions. Nonetheless, the models rely on accurate identification and modeling of the fundamental Markov chain that dictates the variance regimes. This work presents the Large Deviation Markov Switching GARCH model. The large deviation principle is employed to discern the fundamental variance regimes. The concealed states of the heteroskedastic time series are modeled and then employed to fit the Markov switching GARCH model to the heteroskedastic series. The model is subjected to a comparison study with the MSGARCH and standard GARCH models, which are implemented using the MSGARCH and rugarch packages in R, respectively. The in-sample prediction forecasting accuracy of the three models was evaluated using RMSE, MAE, and correlation coefficient metrics. The findings indicated that the proposed model displayed enhanced predictive accuracy relative to the conventional GARCH and MSGARCH models. The LDP’s identification of variance regimes within the series improved its capacity to estimate volatility, which enhances the model’s suitability for volatility modeling.
Received: July 21, 2025
Revised: July 28, 2025
Accepted: September 8, 2025
References
[1] Bashar Yaser Almansour, Muneer M. Alshater and Ammar Yaser Almansour, Performance of ARCH and GARCH models in forecasting cryptocurrency market volatility, Industrial Engineering & Management Systems 20(2) (2021), 130-139.
[2] David Ardia et al., Forecasting risk with Markov-switching GARCH models: a large-scale performance study, International Journal of Forecasting 34(4) (2018), 733-747.
[3] Luc Bauwens, Arie Preminger and Jeroen VK Rombouts, Theory and inference for a Markov switching GARCH model, Econom. J. 13(2) (2010), 218-244.
[4] Jean-Francois Beaumont and Louis-Paul Rivest, Dealing with outliers in survey data, Handbook of Statist. 29 (2009), 247-279.
[5] Giancarlo Corsetti, Paolo Pesenti and Nouriel Roubini, What caused the Asian currency and financial crisis? Japan and the World Economy 11(3) (1999), 305-373.
[6] Manfred Deistler and Wolfgang Scherrer, ARCH and GARCH models, Time Series Models, Springer, 2022, pp. 191-198.
[7] James Ellison et al., The war in Ukraine, Cold War History 23(1) (2023), 121-206.
[8] Robert F. Engle and Tim Bollerslev, Modelling the persistence of conditional variances, Econometric Rev. 5(1) (1986), 1-50.
[9] Ahmed Ghezal and Imane Zemmouri, On the Markov-switching autoregressive stochastic volatility processes, SeMA Journal 81(3) (2024), 413-427.
[10] Gary B. Gorton, Slapped by the Invisible Hand: The Panic of 2007, Oxford University Press, 2010.
[11] Stephen F. Gray, Modeling the conditional distribution of interest rates as a regime-switching process, Journal of Financial Economics 42(1) (1996), 27-62.
[12] Markus Haas, Stefan Mittnik and Marc S. Paolella, A new approach to Markov-switching GARCH models, Journal of Financial Econometrics 2(4) (2004), 493-530.
[13] James D. Hamilton, A new approach to the economic analysis of nonstationary time series and the business cycle, Econometrica 57 (1989), 357-384.
[14] Lokman Kantar, ARCH models and an application on exchange rate volatility: ARCH and GARCH models, Handbook of Research on Emerging Theories, Models and Applications of Financial Econometrics, 2021, pp. 287-300.
[15] Franc Klaassen, Improving GARCH Volatility Forecasts with Regime-switching GARCH, Springer, 2002.
[16] Christopher G. Lamoureux and William D. Lastrapes, Persistence in variance, structural change, and the GARCH model, J. Bus. Econom. Statist. 8(2) (1990), 225-234.
[17] John D. Levendis, Structural breaks, Time Series Econometrics: Learning Through Replication, Springer, 2023, pp. 175-199.
[18] Iana Liadze et al., Economic costs of the Russia-Ukraine war, The World Economy 46(4) (2023), 874-886.
[19] Leandro Maciel, Cryptocurrencies value-at-risk and expected shortfall: do regime-switching volatility models improve forecasting? International Journal of Finance & Economics 26(3) (2021), 4840-4855.
[20] Henri Meier et al., Finanzplatz Schweiz: Finance Center Switzerland, Swiss Finance: Banking, Finance and Digitalization, Springer, 2023, pp. 111-62.
[21] Mamun Miah and Azizur Rahman, Modelling volatility of daily stock returns: is GARCH (1, 1) enough? American Scientific Research Journal for Engineering, Technology and Sciences (ASRJETS) 18(1) (2016), 29-39.
[22] Peterson K. Ozili, Global economic consequences of Russian invasion of Ukraine, Dealing with Regional Conflicts of Global Importance, IGI Global, 2024, pp. 195-223.
[23] V. Pata, Fixed Point Theorems and Applications, UNITEXT, Springer International Publishing, 2019.
url: https://books.google.co.ke/books?id=YQ2xDwAAQBAJ.
[24] Seuk Wai Phoong, Seuk Yen Phoong and Shi Ling Khek, Systematic literature review with bibliometric analysis on Markov switching model: methods and applications, Sage Open 12(2) (2022), 21582440221093062.
[25] Adhya Vagish and Aditya Rao, Standard & Poor’s 500 Index: a trading forecasting analysis through generative artificial intelligence, UC Merced Undergraduate Research Journal 17(1) (2024), 1-71.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Far East Journal of Theoretical Statistics, PUSHPA PUBLISHING HOUSE, PRAYAGRAJ, INDIA
This work is licensed under a Creative Commons Attribution 4.0 International License.
____________________________
Attribution: Credit Pusha Publishing House as the original publisher, including title and author(s) if applicable.
Non-Commercial Use: For non-commercial purposes only. No commercial activities without explicit permission.
No Derivatives: Modifying or creating derivative works not allowed without written permission.
Contact Pusha Publishing House for more info or permissions.
